Chapter 8 and previous parts of the book warned about using the Capital Asset Pricing Model (or any asset pricing model) to estimate the required return (cost of capital). The inputs for the CAPM are just too speculative and value investors avoid building speculation into their valuations.
- The market risk premium, the expected return on the market portfolio minus the risk-free rate is highly speculative. What is the expected return on the market portfolio? Anyone’s guess.
- Beta is estimated with error, with standard errors from estimates from past data about 0.2 to 0.3. So an estimated beta of 1.3 could be 1.0 or 1.6.
- Beta estimated from past data is not the beta going forward during the time an investment is held. Betas change; in technical terms, they are nonstationary.
- The risk-free rate, often set as the yield on a U.S. Treasury bond, in not really risk free. It is at risk from inflation and government deficit spending.
Empirical tests show that betas don’t predict the returns that an investor would expect for risk. Using the CAPM is playing with mirrors. “Beta is dead.”
Several surveys do provide consensus estimates of the required return. For example, see:
Fernandez, P., D. Garcia de la Garza, and P. Acin. 2023. Survey: Market Risk Premium and Risk-Free Rate Used for 80 Countries in 2023. At https://ssrn.com/abstract=4407839.
This survey is updated each year.
Duff & Phelps Valuation Handbook-U.S. Industry Cost of Capital. Maintained by Kroll. See https://www.kroll.com/-/media/kroll-images/pdfs/webcast-cost-of-capital-in-the-current-environment.pdf.
James Harrington and others, Valuation Handbook-International Guide to the Cost of Capital, CFA Institute Foundation. At https://www.cfainstitute.org/-/media/documents/book/rf-publication/2021/igcc-summary-edition-2021.pdf. This is the 2021 edition.
The consensus estimate of the market risk premium typically comes in at about 5.5%. This happens to be the number that professors in business schools tell students to use, perhaps just parroted.
Here’s a video on the standard calculation of the cost of capital. As you watch it ask: Is this any good? Can we get a handle with these calculations going forward? Are we guessing?